The influence of parental financial teaching on saving and ... · Thesis, MSc. Finance University...
Transcript of The influence of parental financial teaching on saving and ... · Thesis, MSc. Finance University...
The influence of parental financial teaching on saving and borrowing
behavior
Annewil Mettien Homan1
Thesis, MSc. Finance
University of Groningen
Supervisor: Dr. M.M. Kramer
January 2016
12,900 words
Abstract
This paper uses the household survey of De Nederlandsche Bank to investigate whether
parental financial teaching influences the saving and borrowing behavior of individuals. The
findings are that parental financial teaching leads to more saving and less borrowing. In
addition, the study shows that this teaching most affects saving behavior when parents
provide it throughout the childhood period. It also indicates that modeling, guidance and
discussion, and habit formation mechanisms explain the influence parents have on the
formation of their children’s saving and borrowing behavior.
Keywords: household financial decision making, parental financial teaching, borrowing,
saving
JEL classifications: A290, D120, D140, D830, J130
1 University of Groningen, Faculty of Economics and Business
Student number: s2196034 E-mail: [email protected]
The influence of parental financial teaching on saving and borrowing behavior
2
1. Introduction
Households nowadays have more responsibilities in relation to financial issues; in
addition, they also have to make an increasing number of decisions on their own. For
instance, government policies have made pensions subject to less government regulation and
more the responsibility of individuals. Households also have easier access to financial
markets and increasing possibilities for obtaining credit. However, the exposure to risk has
grown as a result of changes to the rules for and terms of borrowing money in recent years
(Lusardi and Mitchell, 2014). These changes in the financial environment have led to
financial decision making becoming an active topic for households, which must now decide
everything from how much money to put aside in the following year to whether to take out
loans.
Recent studies have shown that households struggle to make responsible financial
decisions and therefore display suboptimal behavior. For instance, households fail to put
enough money aside for retirement (Lusardi and Mitchell, 2007b), four out of ten households
in the Netherlands have insufficient funds to deal with financial setbacks (Boonen, 2015).
Households in the United States have, on average, €7,500 of uncollateralized debt (Benton,
Meier and Sprenger, 2007).
A major reason why households make suboptimal financial decisions is that they have
a limited degree of financial literacy. Lusardi and Mitchell (2014) describe financial literacy
as “peoples’ ability to process economic information and make informed decisions about
financial planning, wealth accumulation, debt, and pensions.” Financial illiteracy leads to
lower savings and more borrowing (Alessie, van Rooij and Lusardi, 2011). Increasing the
financial literacy of households is therefore a solution to the problem of suboptimal behavior.
One of the factors that may positively influence financial literacy levels is financial
education (Bucks and Pence, 2006; Batty, Collins and Odders-White, 2015). Studies that have
looked at topics ranging from financial education at high school levels (Bernheim, Garrett and
Maki, 2000) to the effects of financial education in the workplace (Bayer, Bernheim and
Scholz, 2009) conclude that financial education leads to changes in financial behavior.
However, the effect of financial education by parents has not been widely studied, despite that
several studies indicate that the financial behavior that is formed in childhood, which likely
partly stems from one’s parents, persists into adulthood (Shim, Barber, Card, Xiao and Serido,
2010). This study therefore investigates the role of parents in the formation of their children’s
saving and borrowing behavior.
The influence of parental financial teaching on saving and borrowing behavior
3
Although this study is not the first to look at the effect of parental financial teaching
on saving and borrowing behavior, the understanding of this effect is still in its infancy. This
research also constitutes a valuable contribution to existing research as it investigates the
mechanisms behind parental influence on children’s future saving and borrowing behavior
and consequently better explains parents’ role in the process. Additionally, this study reports
whether the age at which children learn about financials from their parents makes a difference
and clarifies at which age the formation of individual saving and borrowing behavior is most
affected by parents. It also examines the long-term effects of receiving parental financial
teaching, which is a first in relation to borrowing behavior (given that previous studies have
only observed the effect on young adults). The findings of this research are therefore more
representative in terms of explaining the actual saving and borrowing behavior of adults.
This research explores saving and borrowing behavior, since saving and borrowing
decisions are the most important financial decisions that people make. Consumers save and
borrow to have enough money left for retirement, to make stable and expensive purchases and
to protect themselves against unforeseen expenses (Bucciol and Veronesi, 2014). Although
these motives are positive, borrowing money also has adverse effects, given that over-
borrowing can put households into financial difficulties. It is therefore useful to research
whether parental financial teaching can lead to specific consumer saving and borrowing
behavior. As such, the research question of this study is as follows:
Does parental financial teaching in childhood influence individual saving and
borrowing behavior?
Answering this question reveals whether the “financial education” method that parents use
with their children influences the saving and borrowing behavior of consumers. I do so using
panel data from the De Nederlandsche Bank (DNB) household survey (DHS) that includes
information obtained during annual interviews with Dutch households on matters pertaining
to household finances.
My empirical analysis produces several results. First, I find that financial parental
teaching leads to 2% more willingness to save and 1% fewer loans. Second, I discover that it
also leads to more savings and lower debts. Third, my results show that providing parental
financial teaching over a long period in childhood influences saving behavior most
significantly. Fourth, I find that modeling, guidance and discussion, and habit formation
mechanisms explain parental effects on children’s future saving and borrowing behavior.
The influence of parental financial teaching on saving and borrowing behavior
4
This study is relevant for both policy makers and academics. Policy makers want
individuals to make good financial decisions, and this research reveals how parents influence
this achievement. The relevance for academics is twofold: first, this research links to previous
studies on the impact of financial education on saving and borrowing behavior. Second, to my
knowledge it is one of the first studies to observe different mechanisms in order to explain the
effect of parental financial teaching on the saving and borrowing behavior of adults. This
research is therefore a valuable addition to the existing literature.
This paper is organized as follows: section two provides an overview of previous
studies on this topic; section three contains a description of the data and the methodology;
section four outlines the main empirical findings; and finally, section five presents
conclusions and discusses the research.
2. Parental financial teaching: Theoretical background and hypothesis setting
This section provides the theoretical background on parental financial teaching and
individuals’ saving and borrowing behavior. A literature review has been performed to
formulate the hypotheses.
2.1. The influence of parents
Many previous studies have discussed the role that financial education plays in
relation to financial behavior and whether financial education is relevant to increasing
financial knowledge (Bucks and Pence, 2006; Batty et al., 2015). However, it is difficult to
make education that is suitable for everyone (Willis, 2011; Lusardi and Mitchell, 2014), as
some people have more basic knowledge about financials than others. Consequently, financial
education in schools or at the workplace is not the most significant factor that affects the
saving and borrowing behavior of adults. In line with this, Shim et al. (2010) argue that the
role that parents play in predicting young adult behavior is substantially greater than the role
played by work experience and high school financial education, since the financial behavior
formed in childhood persists into adulthood. Additionally, social and familial influences result
in particular financial behavior before children are formally educated (Batty et al., 2015). It is
therefore likely that parental financial teaching is more appropriate and effective than general
financial education. Vassallo (2003) supports this view by stating that parents are the most
The influence of parental financial teaching on saving and borrowing behavior
5
important socialization agents. The theory of planned behavior and the family systems theory
both support the view that parents play a role in forming the financial behavior of their
children; these theories are further explored below.
First, the theory of planned behavior states that an individual’s behavior is influenced
by behavioral intentions that are formed by attitudes, subjective norms and perceived control
(Ajzen, 1991). Parents influence children’s attitude towards particular behaviors, since
children are likely to copy some of their parents’ behaviors. In addition, children feel pressure
from their parents to behave in a certain way (e.g. not to spend much money on candy).
Finally, parents influence perceived control, since they can control their children’s
expenditures (e.g. by prohibiting certain purchases). This theory therefore asserts that parents
influence their children’s behavioral intentions. According to Ajzen (1991), the stronger the
intention to engage in a behavior, the more likely it is that they behavior will be exhibited. It
is correspondingly expected that parents influence the saving and borrowing behavior of their
children.
Second, the family systems theory also suggests that parents influence the behavior of
their children. The theory is derived from Bertalanffy’s (1968) systems theory, which states
that the behaviors of people in the same group (system) are correlated with each other. In
relation to families, this means that the individual actions of family members affect the
behavior of the whole family (Moore and Asay, 2013). The family systems theory
consequently supports the assertion that parents’ actions (e.g. monitoring expenditure or
giving advice) influence the behavior of their children.
Previous studies thus reveal that a relationship between parents and the future saving
and borrowing behavior of children does exist. Furthermore, Bucciol and Veronesi (2014)
find that parental financial teaching increases an individual’s willingness to save by 16% and
their saving amounts by about 30%, while Norvilitis and MacLean (2010) find that parents
influence the borrowing behavior of adults. The first hypothesis is therefore as follows:
Hypothesis 1a: Receiving parental financial teaching influences the saving behavior of
adults.
Hypothesis 1b: Receiving parental financing teaching influences the borrowing
behavior of adults.
The influence of parental financial teaching on saving and borrowing behavior
6
2.2. The influence of prolonged education
According to Bayer et al. (2009), the greatest effect on financial behavior is achieved
by receiving financial education as often as possible. They find that for employees, attending
more seminars leads to better financial behavior. When their findings are connected with this
study, it can be expected that receiving more parental financial teaching in life leads to major
changes in saving and borrowing behavior. Parents can teach their children about financial
behavior at different ages: parental financial teaching can begin in early childhood and
continue through adolescence. Not much research has been done on the ages at which children
can best receive parental financial teaching; however, it is likely that prolonged financial
education is the best, since children have more frequent contact with the advantages of
parental financial teaching, which leads to a more persistent effect. Moreover, some financial
issues are too hard to discuss at a young age and should be deferred to later. However, starting
financial teaching too late can mean that unwanted habits have already been formed.
Providing parental financial teaching throughout childhood should therefore lead to the best
results. This is in line with the findings of Bucciol and Veronesi (2014), who suggest that the
most effective way to influence individuals’ saving behavior is to teach them during both
childhood and adolescence. The second hypothesis is therefore as follows:
Hypothesis 2a: Receiving parental financial teaching during both childhood and
adolescence has the largest impact on the saving behavior of adults.
Hypothesis 2b: Receiving parental financial teaching during both childhood and
adolescence has the largest impact on the borrowing behavior of adults.
2.3. An overview of different mechanisms
As stated in the introduction, another interesting issue to explore is how parents
influence the saving and borrowing behavior of their children. To do so I test four
mechanisms introduced by Webley and Nyhus (2006), namely modeling, discussion and
guidance, habit formation, and independence.
The first mechanism is modeling. In general, parents are the first models observed by
children; children imitate their behavior or compare it with their own. This is in line with the
findings of Webley and Nyhus (2006), who discovered that the approach of children to
The influence of parental financial teaching on saving and borrowing behavior
7
economic matters is linked to that of their parents. It is consequently expected that when
children observe particular behavior in their parents (e.g. the careful use of money), they will
copy it; it is also possible that they show contrarian behavior, since some children turn away
from their parents. Even in the latter case, however, as role models parents still do influence
their children’s future saving and borrowing behavior.
Discussion and guidance constitute the second mechanism that can explain the relation
between parental financial teaching and adult financial behavior. Related activities include
explaining financial situations to children, discussing why certain financial decisions are
made in households and providing advice about saving and borrowing. Children who discuss
financial issues with their parents are stimulated to think about financial matters, which
increases their financial awareness. In addition, Webley and Nyhus (2006) state that
individuals whose parents discuss household economic decisions with them have both a better
future orientation and an ability to control expenditure. The former should lead to higher
savings (with a view to retirement), while the latter leads to both higher savings (as less
expenditure means more money is left to save) and less over-borrowing (if consumers can
better control their expenditures, fewer loans are need to cover expenses). Discussing
financial matters with and providing guidance to children are therefore expected to influence
saving and borrowing behavior. Previous studies support this view: Bucciol and Veronesi
(2014) find that receiving advice from parents leads to both a higher propensity to save and
higher saving amounts, while Norvilitis and MacLean (2010) find that teaching children how
to manage their allowance and bank accounts leads to lower credit card debt levels in college.
The third mechanism is habit formation. As individuals’ habits are formed at a young
age, parents influence the development of their children’s habits. According to Batty et al.
(2015), schools cannot fill the role of parents in relation to personal finance habits. Parents’
role in habit formation thus has a high impact. For instance, the way a source of income is
managed is a habit that can be formed in childhood. Examples of habits are spending money
immediately or saving it for the future. Parents influence the habit formation of their children
by monitoring their spending patterns and pushing behavior in certain directions to prevent
unwanted habits from being formed. As it is likely that behavior from childhood persists into
adulthood, parents’ role in the formation of their children’s habits allows them to influence
their saving and borrowing behavior as well.
Independence is the fourth mechanism that can explain the role of parents in forming
the future behavior of children. Independence in this case means that children are not
dependent on their parents in order to make some expenditures. One method for creating
The influence of parental financial teaching on saving and borrowing behavior
8
independence is giving children an allowance, which makes children responsible for
managing their own money. This teaches children to make their own decisions, which leads to
experience in making financial decisions (Ashby, Schoon and Webley, 2011). In addition,
Webley and Nyhus (2006) argue that having responsibility for financial decisions leads to
learning economic skills. Since independence in making financial decisions in childhood
increases both the experience in making financial decisions as well economic skills, it is
expected to influence the future saving and borrowing behavior.
Beyond the four mechanisms of Webley and Nyhus (2006) described above,
heritability also plays a role in the formation of saving and borrowing behavior. Cronqvist and
Siegel (2015) find that 33% of the variation in savings behavior across individuals can be
explained by genetic differences. In addition, Nyhus and Webley (2001) have observed that
personal traits influence saving and borrowing behavior. Personal traits are inherited and
therefore heritability can be one explanation as to why parents influence the future saving and
borrowing behavior of their children. However, as the dataset used in this study does not
cover heritability, the effect of heredity cannot be investigated.
Considering the four mechanisms discussed above, the third hypothesis is as follows:
Hypothesis 3a: Modeling, discussion and guidance, habit formation and independence
are mechanisms that explain how parents influence the saving behavior of adults.
Hypothesis 3b: Modeling, discussion and guidance, habit formation and independence
are mechanisms that explain how parents influence the borrowing behavior of adults.
3. Data and methodology
This section first provides an overview of the data source and sample selection method.
Second, the different variables used in this study are described. Finally, the methodology is
discussed.
3.1. Data selection
This study uses panel data from the DHS to test whether parental financial teaching
influences the financial behavior of individuals. The survey has been conducted by
The influence of parental financial teaching on saving and borrowing behavior
9
CenterERdata every year since 1993. The information is collected to explain the influences of
economic and psychological factors on the financial behavior of Dutch households.
According to Webley and Nyhus (2006), the sample is representative of the Dutch population.
The DHS data are collected by means of an Internet panel undertaken by CentERdata.
People without connectivity participate through a TV set top box or using an obtained
computer and internet connection. Each year, approximately 2,000 households participate in
the survey. All persons over the age of sixteen in the household are invited to complete it. In
order to deal with attrition, households in the dataset are replaced over the years (Teppa and
Vis, 2012). The survey gathers information on participants’ saving and borrowing behaviors
and childhoods, as well as general information.
An advantage of the DHS is that it tracks the same individual over a number of years,
which allows researchers to obtain a representative picture of respondents’ financial behavior.
Moreover, the influence of factors that fluctuate over the years is reduced.
This research considers surveys from the years 2000 to 2014. Surveys before 2000 are
not included because, according Bucciol and Veronesi (2014), they were primarily completed
by wealthier households. Furthermore, after 2000 the questionnaires are similar and the
sampling design is the same (Bucciol and Veronesi, 2014). Consequently, the surveys from
2000 onwards are more comparable and the results are easier to interpret.
The original dataset retrieved from the DHS contained 69,168 observations. From this
dataset, 18,838 observations are used in the final sample for this research. First, 40,725
observations are deleted because the respondent was not the head of the household. According
to Nyhus and Webley (2001), the head of the household has the most influence on how much
to save and to borrow. Moreover, only the person who pays the accounts responds to
questions in the DHS about shared savings accounts and loans, and this person is often the
head of the household; as such, only answers from heads of households are relevant. Second,
in order to observe the influence of parental financial teaching on adults, seven observations
from respondents younger than seventeen are not included in the study. Third, 8,834
observations are deleted from the original dataset because the respondents did not answer a
question on parental financing teaching or respond to one of the dependent variables, which
makes it impossible to investigate the influence of financial parental financial teaching on
saving and borrowing behavior. A further 677 observations are deleted because the
individuals only participated in the survey on one occasion; repeated observations from the
same individuals are needed to produce better estimates for the time-varying characteristics of
the respondents.
The influence of parental financial teaching on saving and borrowing behavior
10
The questions about financial parental financial teaching relate to a respondent’s past;
as such, the responses given should be the same from year to year. However, this was not
always the case. In order to obtain consistency, inconsistent answers are replaced with the
prevailing answer of the respondent over the years. It is likely that the answer the respondent
provides most commonly is the correct one. Respondents who do not have a prevailing
answer are deleted.
After all of the above considerations are taken into account, the result is a final sample
of 18,838 observations. These observations come from 2,754 individuals. Respondents are
always the head of the household, so the 18,838 observations are from 2,754 unique
households. The average number of observations per respondent is 7.3, which is sufficient to
reveal patterns in respondents’ saving and borrowing behaviors and investigate factors that
might explain them.
Table I presents summary statistics for the pooled sample. The average age of
respondents is 54 and approximately 25% of the individuals are retired. Men are over-
represented in this research, since only 25% of the respondents are women. Slightly more than
half of the respondents are employees. Seventy percent of the respondents save an average
amount of €4,116 per year (which represents on average 13% of their net income). In
comparison, 22% of the respondents have loans with an average debt shield of €3,449 (which
is on average 9% of net income). Almost 85% of the respondents received parental financial
teaching in both their young childhood and adolescence.
The influence of parental financial teaching on saving and borrowing behavior
11
Table 1. Summary statistics - pooled sample
This table provides an overview of summary statistics of the pooled sample. Saving behavior refers to four variables which measure the saving behavior of individuals. Debt behavior refers to three variables which measure the debt behavior of individuals. Parental teaching
received in childhood refers to the degree in which parental financial teaching is obtained in childhood. Parental teaching only received
in young childhood, parental teaching only received in adolescence and parental teaching received throughout childhood are dummy variables equal to 1 if parental financial teaching is only received in the corresponding age category. Presence of mechanisms in
childhood refers to four ordinal variables which indicate in which degree the different mechanisms were present in childhood. Control
variables refers to socio-demographic characteristics of the respondents.
Variable Range Obs. Mean Std. Dev.
Saving behavior
Willingness to save 0 - 1
17,273
0.70
0.46
Total savings per year (€)
16,637
4,116
7,328
Willingness to save in future 1 - 4
16,943
3.23
0.87
Savings to net income
11,607
0.13
0.71
Debt behavior
Presence of a loan 0 - 1
16,500
0.22
0.42
Total debt (€)
12,747
3,449
33,914
Total debt to net income
8,987
0.09
1.61
Parental financial teaching
Parental financial teaching received 4 - 17
17,019
11.16
2.55
Parental teaching only received in young childhood 0 - 1
18,420
0.13
0.33
Parental teaching only received in adolescence 0 - 1
18,420
0.01
0.10
Parental teaching received throughout childhood 0 - 1
18,420
0.83
0.37
Presence of mechanisms in childhood
Modeling / received stimulation to save 1 - 4
18,150
2.67
1.05
Discussion and guidance / received financial advice 1 - 4
18,039
2.45
1.09
Habit formation / expenditures were monitored 1 - 5
18,022
3.44
1.36
Independence / received pocket money 1 - 4
18,443
2.58
1.36
Control variables
Female 0 - 1
18,838
0.25
0.43
Age (in years)
18,837
54
14.92
Retired (age >65) 0 - 1
18,838
0.27
0.45
High education 0 - 1
17,301
0.44
0.50
Middle education 0 - 1
17,301
0.52
0.50
Low education 0 - 1
17,301
0.04
0.20
Employee 0 - 1
16,498
0.58
0.49
Self-employee 0 - 1
16,498
0.04
0.19
Participation in a partnership 0 - 1
16,498
0.01
0.07
No work 0 - 1
16,498
0.38
0.49
Risk attitude 1 1 - 7
15,900
4.96
1.37
Risk attitude 2 1 - 7
15,933
5.47
1.25
Number of children 0 - 6
18,838
0.59
1.01
Size of the household 1 - 8
18,838
2.27
1.24
Net income (€) 12,587 32,005 21,725
The influence of parental financial teaching on saving and borrowing behavior
12
3.2. Construction of key variables
Below the independent, dependent and control variables used in this study are
described successively.
3.2.1. Saving and borrowing behavior
The dependent variables in this research are the saving and borrowing behaviors of
adults. Please refer to appendix A for an exact definition of all of the variables included.
Saving behavior is measured in four ways. First, as the willingness to save in a given
year; second, as the total amount of savings in a given year; third, as the ratio of savings in a
given year relative to net income; and fourth, as the propensity to save in the future. The
willingness to save takes a value of one (yes) if the respondent’s household put any money
aside during the previous year; if not, it has the value of zero (no). This is in line with the
studies of Webley and Nyhus (2012) and Bucciol and Veronesi (2014). The amount saved in
the previous year is derived from a question with a discrete answer scale. I follow the
approach of Bucciol and Veronesi (2014) and include this as a continuous variable. The
continuous variable is set equal to the central value of each range. For the extreme ranges, the
threshold value is used to create a continuous variable. In addition, the natural logarithm of
the amount saved in a given year is included in the study in order to reduce the influence of
outliers on its results. The amount of savings increases over time in line with inflation, which
represents no real increase; it should therefore be corrected for inflation. The consumer price
index for all items is used2 to report savings at 2013 prices. The ratio of savings in a given
year relative to net income is measured as the sum of total savings in a given year divided by
net income of a given year. The propensity of individuals to save in the future is measured by
the question: “Is your household planning to put money aside in the next 12 months?” The
answer is included as an ordinal variable [1 = No, certainly not; 2 = No, probably not; 3 =
Yes, perhaps; 4 = Yes, certainly]. These measurements of future saving behavior correspond
with the studies of Nyhus and Webley (2001) and Webley and Nyhus (2012).
To test the borrowing behavior of households, eight types of loans are distinguished in
this research: private loans; extended lines of credit; credits by mail-order companies; finance
2 Source: Statistics Netherlands, the consumer price index can be found at http://statline.cbs.nl/StatWeb/publication/?DM=SLNL&PA=71905NED&D1=a&D2=a&VW=T
The influence of parental financial teaching on saving and borrowing behavior
13
credit; loans from family, friends, or acquaintances; study loans; credit cards; and other loans.
This is in line with the studies of Nyhus and Webley (2001) and Webley and Nyhus (2012).
Borrowing behavior is measured in three ways: first, as the presence of a loan;
second, as the total sum of debt on different loans; and third, as the ratio of the total amount
of debt relative to net income. The presence of a loan is included as a dummy variable in the
study (0 = No; 1 = Yes). The variable takes a value of one if the respondent has one or more
loans as described above, otherwise it has a value of zero. The total amount of debt is
calculated as the sum of the debt on the different loans. The natural logarithm of the sum is
taken to minimize the effect of outliers. As in the case of the saving amount, this variable is
corrected for inflation. In addition, the total amount of debt for years before 2003 is converted
to euro from guilder, which makes the variable comparable. The portion of debt that a person
has relative to his or her net income is measured by the sum of the total amount of debt
divided by net income.
3.2.2. Parental financial teaching
This study uses four questions from the DHS to observe whether an individual
receives parental financial teaching. Please refer to appendix A for more information on
these questions. In this research, the parental financial variable is the total sum of the ordinal
answer scales for these four questions, which range from 4 to 17. For clarity, a high score on
this variable means that the respondent received a great amount of financial education from
his or her parents and the minimum score of 4 refers to an individual who has not received
parental teaching in their childhood. This measurement was chosen given that more than 95%
of the respondents have received some form of parental financial teaching. A dummy variable
is therefore not suitable for detecting anything, since hardly any variance in that variable
would be seen.
The four questions about the presence of parental financial teaching are also used to
observe the effect of possible mechanisms that can explain parental influence on the future
saving and borrowing behavior of children. First, being stimulated to save in childhood is
used as a proxy for modeling. Although it is does not directly measure the role of modeling
by parents, it measures its indirect effects. I assume that parents stimulate their children to
mimic their financial behavior by emphasizing that behavior; in other words, parents provide
an example of how to behave in relation to financial issues. Coming into contact with this
behavior makes it easier for children to copy it. Since this copying is an effect of modeling,
The influence of parental financial teaching on saving and borrowing behavior
14
the stimulation of particular behavior is linked with parents as models. The following question
is selected as the best option for measuring modeling using the DHS dataset: “Did your
(grand)parents stimulate you to save money between the ages of 12 and 16?” The variables
range from 1 to 4 [1 = No, not at all; 2 = Yes, but only to a certain extent; 3 = Yes, they told
me how important saving is; 4 = Yes, they emphasized the necessity of saving].
Second, discussion and guidance are measured by looking at budget advice received
from parents during childhood. The following question is used: “Did your (grand)parents try
to teach you how to budget when you were between 12 and 16 years of age?” This variable
also ranges from 1 to 4 [1 = No; 2 = Yes, but only to a certain extent; 3 = Yes, they gave me
some advice and practical help; 4 = Yes, they gave me advice and practical help].
Third, parental monitoring is used as a proxy for habit formation. As mentioned in
Section 2.3, monitoring expenditures is a method for parents to influence the formation of
habits in their children. “When you were between 8 and 12 years of age, could you spend your
money as you pleased?” is the question used. This variable ranges from 1 to 5 [1 = I could
decide on all of my expenditures; 2 = I could mostly decide how I spent my money; 3 = Part
of my expenditure was decided by me, the rest by my parents; 4 = My parents decided how I
spent most of my money; 5 = My parents decided how I spent all of my money].
Lastly, receiving pocket money from parents is used as proxy for independence. As
described in Section 2.3, giving an allowance is one way to make children financially
independent and responsible. The following question is used: “When you were between 8 and
12 years of age, did you receive an allowance from your parents? By allowance we mean a
fixed amount received on a regular basis.” The variables range from 1 to 4 [1 = No; 2 =
Occasionally; 3 = Yes, but it was sometimes forgotten; 4 = Yes].
The mechanisms described above are used to measure in which age categories parental
financial teaching is received. First, parental financial teaching in young childhood is
included as dummy variable. It has a value of 1 if parental financial teaching is only received
in young childhood (in other words, only if the mechanisms of habit formation and/or
independence are present and not modeling or/and discussion and guidance). This is because
the questions used for the habit formation and independence mechanisms concern the younger
years (ages 8-12) of the respondents and the questions to detect modeling and discussion and
guidance relate to their adolescence (ages 12-16). Second, parental financial teaching in
adolescence is included as a dummy variable. It has a value of one if parental financial
teaching is only received in adolescence (in other words, only if the mechanisms of modeling
and/or discussion and guidance are present and not habit formation and/or independence).
The influence of parental financial teaching on saving and borrowing behavior
15
Lastly, parental financial teaching throughout childhood is also included as a dummy
variable. It takes the value of one if parental financial teaching is received in both age
categories. A mechanism is present if the value of its corresponding variable differs from one.
The limitation of this method is that no distinction is made between the degrees in which the
mechanisms are present. However, in order to maintain the interpretability of the results a
dummy variable was used. These measurement method is in line with those of Bucciol and
Veronesi (2014). Another caveat is that it does not exclude the presence of the other
mechanisms in the observed age group, since the data do not provide exact information about
the presence of all mechanisms throughout childhood. For instance, a child can also receive
pocket money (a proxy for independence) from ages 12 to 16, which can also have an impact
on their saving and borrowing behavior in adulthood. In addition, the results of this method
can be biased since the observed effect may arise from the functioning of the mechanisms and
not through the age category in which the teaching is received. Despite the limitations, this
method, using the DHS dataset, is the most convenient to conclude anything about the age at
which children can best receive parental financial teaching. Please refer to appendix A for the
exact wording of the survey questions.
3.2.3. Control variables
Many studies have yielded results that pertain to saving and borrowing behavior.
According to Hira and Mugenda (2000), women are more likely than men to make
unnecessary expenditures, which could lead to them having lower savings. A study of Bucciol
and Veronesi (2014) supports this view: its findings indicate that women have lower saving
amounts than men, as well as that individuals who are employed, self-employed or retired
have a higher willingness to save and higher saving amounts than people who do not work.
According to the standard life-cycle model, it is expected that young adults have a lower
willingness to save and that saving amounts are lower for older individuals. The findings of
Bucciol and Veronesi (2014) support this model. According to Webley and Nyhus (2012), the
likelihood that an adult has debt increases with age, however this result is not supported by
the Nyhus and Webley’s study (2001). Next to that, Nyhus and Webley (2001) find that
individuals with high education levels have higher savings and debts. The possible
explanation they offer is that educated people have student loans. By contrast, the research of
Webley and Nyhus (2012) find no significant relation between education and the sum of debt.
The influence of parental financial teaching on saving and borrowing behavior
16
Household characteristics also influence the saving and borrowing behavior of
individuals. Studies of Bucciol and Veronesi (2014) and Nyhus and Webley (2001) find that
larger households have a lower willingness to save and lower saving amounts. These studies
also find that households with higher net income have higher savings amounts. In addition,
Nyhus and Webley (2001) find that households with higher income have more debts, as it is
more likely that they will be granted higher loans than low-income families. However, the
study of Webley and Nyhus (2012) shows no significant relation between net income and the
sum of debt.
Furthermore, according to Jacobs-Lawson and Hershey (2005), risk-tolerant
individuals prefer to invest in risky investments. Since a bank account is not very risky, adults
who are risk-tolerant are expected to have lower savings.
Based on the above discussion, the following socio-demographic variables are
included as controls in the multivariate analysis of this research: gender, occupation, age,
education, household characteristics, net income of household and risk attitude. Please refer to
appendix A for an exact definition of all of the control variables included.
3.3. Methodology
It is likely that changes in economic conditions may lead households to save more in
some years. Furthermore, according to Statistics Netherlands,3 households in the three largest
cities of the Netherlands often live in poverty. It is therefore expected that individuals living
in Amsterdam, Rotterdam and The Hague have lower savings and a higher amount of debt.
Time and area dummy variables, which represent time and area fixed effects, are
consequently included in the estimation, in accordance with the research of Bucciol and
Veronesi (2014).
A generalized least square random effects model is used to test the influence of
receiving parental financial teaching on financial behavior. The results are easier to interpret
than those obtained using probit and ordered probit models.4
A random effect model is provided above ordinary least squares, since the results of
Breusch and Pagan’s (1980) Lagrange Multiplier Test are significant for each regression
model. The Breusch-Pagan statistics test whether the variances across entities are zero. If they
are, no significant difference exists across units (in other words, no panel-effect is present)
3 Source: http://www.cbs.nl/NR/rdonlyres/83CC1C97-DC64-4FEC-8535-C651EC5A62DD/0/armoedesignalement2014pub.pdf 4 Probit and ordered probit random effects models are also conducted for the binary dependent variables. As the results are in line with the findings of the GLS, only the results of the GLS are presented in this study.
The influence of parental financial teaching on saving and borrowing behavior
17
and a simple OLS estimation is desirable (Torres-Reyna, 2007). As a significant difference
across individuals is found in this study, a random effect model is used.
The main assumption of the random effect model is that the time-varying and time-
invariant independent variables are assumed exogenous with respect to the error term.
However, there is an omitted variable bias if this is not true, which means that the coefficients
estimated by a random effect model might be biased (Bucciol and Veronesi, 2014). This can
be solved by estimating the model with fixed effects. However, as the independent variable in
this study is a variable that does not change over time, a fixed-effects model cannot be
estimated. A random effect model is consequently used to conduct the different analyses. In
order to reduce the omitted variable bias, as many control variables as possible are included.
Robust standard errors are used in this study to prevent biased standard errors. This is
in line with the conclusions of Petersen (2009), who states that the best solution for producing
standard errors and correcting confidence intervals is using estimates that are robust to the
form of dependence in the data.
In summary, to test the effect of receiving parental financial teaching on saving and
borrowing behavior, this study estimates the following regression equation:
𝑌𝑖,𝑡 = 𝛼 + 𝑃𝑎𝑟𝑒𝑛𝑡𝑎𝑙 𝑓𝑖𝑛𝑎𝑛𝑐𝑖𝑎𝑙 𝑡𝑒𝑎𝑐ℎ𝑖𝑛𝑔 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒𝑠 𝑖 ∗ 𝛽 + 𝐶𝑜𝑛𝑡𝑟𝑜𝑙 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒𝑠𝑖𝑡 ∗ 𝛽
+ 𝑇𝑖𝑚𝑒 𝑎𝑛𝑑 𝑎𝑟𝑒𝑎 𝑑𝑢𝑚𝑚𝑦 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒𝑠𝑖𝑡 ∗ 𝛽 + 𝜔𝑖𝑡 , 𝜔𝑖𝑡 = 𝜇𝑖 + 𝜀𝑖𝑡 (1)
Where 𝑌𝑖,𝑡 = 𝑙𝑎𝑡𝑒𝑛𝑡 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒, 𝑖 = 𝑟𝑒𝑠𝑝𝑜𝑛𝑑𝑒𝑛𝑡, 𝑡 = 𝑦𝑒𝑎𝑟 𝑎𝑛𝑑 𝜔𝑖,𝑡 𝑖𝑠 𝑡ℎ𝑒 𝑒𝑟𝑟𝑜𝑟 𝑡𝑒𝑟𝑚. The
error term consists of a cross-sectional error term (𝜇𝑖) and an individual observation error
term (𝜀𝑖𝑡). In this study, the latent variable is replaced with two different dependent variables:
saving behavior and borrowing behavior. The parental financial teaching variable is a vector
of parental financial teaching variables and the control variable is a vector of the control
variables. Time and area dummy variables respectively represent the year of the survey and
the area where the respondent lives.
4. Results
The results of this study concerning parental financial teaching on saving and
borrowing behavior are presented below in two categories: univariate and multivariate. The
The influence of parental financial teaching on saving and borrowing behavior
18
results of an additional test undertaken to explore the effects of parental financial teaching on
borrowing behavior is also presented.
4.1. Univariate results
Table II provides an overview of the effects that parental financial teaching in
childhood, parental financial teaching in different age categories and different mechanisms
have on facets of saving and borrowing behavior.
Panel A shows that parental financial teaching leads to significant changes in both
saving and borrowing behavior, as per the ANOVA statistics. The parental financial teaching
variable is split into quartiles to make the results easier to interpret. The willingness to save is
18% higher in the quartile of individuals who received the most parental financial teaching in
comparison to the lowest quartile. Adults in the highest quartile have approximately €1,500
more savings compared to individuals who received little or no parental financial teaching.
The willingness to put money aside in the future is also most present in the highest quartile.
Individuals who received much parental financial teaching have 4% fewer loans than those in
the lowest quartile. Since the amount of debt is the lowest in the third quartile group, this
finding is the first indication that the degree of parental financial teaching has an impact on
saving and borrowing behavior.
Panel B then proves that only parental financial teaching during young childhood leads
to more savings and fewer loans in adulthood compared with parental financial teaching
during adolescence. However, adults who received parental financial teaching in young
childhood and adolescence have the highest savings and smallest presence of loans. By
contrast, adults have the lowest amount of debt when they only received parental teaching
during adolescence. However, this difference is not significant according the ANOVA test.
These results therefore indicate that providing parental financial teaching over a long time has
the greatest influence on the saving and borrowing behavior of individuals.
Next, panel C indicates that stimulating children to save leads to higher willingness to
save and greater savings in adulthood. Adults who were stimulated to start saving most often
save almost €1,300 more than those who were not encouraged to save during childhood.
Moreover, being stimulated to save leads to lower numbers of both loans and debts. This is
reasonable, since stimulating to save in childhood leads to higher savings and thus less need
to borrow.
The influence of parental financial teaching on saving and borrowing behavior
19
Table II: Univariate results
This table presents the univariate results of the effects that parental financial teaching in childhood (panel A), parental financial teaching in different age categories (panel B) and different mechanisms (panel C-F) have on the following facets of
saving and borrowing behavior: willingness in the past to save (column 1), amount of savings per year (column 2), willingness to save in future (3), presence of a loan (4), and total amount of debt (column 5). *,**,*** denote significance on the
ANOVA test at the 10%, 5% and 1% level respectively.
Willingness to
save
Total savings
per year
Willingness to
save in future
Presence of a
loan Total debt
Willingness to
save Total savings
Willingness to
save in future
Presence of a
loan Total debt
(1) (2) (3) (4) (5) (1) (2) (3) (4) (5)
Panel A: Parental financial teaching received (divided in quartiles)
Panel D: Presence of discussion/guidance as mechanism (received financial advice)
1 (low) 0.61 3,464 3.06 0.23 3,128
No 0.62 3,737 3.06 0.22 3,717
2 0.70 4,129 3.21 0.22 3,719
Yes, but only to a certain
extent 0.70 3,985 3.23 0.21 2,969
3 0.70 4,158 3.24 0.23 2,899
Yes, they gave me some
advice and practical help 0.71 4,215 3.26 0.23 3,036
4 (high) 0.79 5,087 3.44 0.19 3,662
Yes, they gave me advice
and practical help 0.76 4,633 3.37 0.20 4,046
P-value ANOVA statistic 0.00*** 0.00*** 0.00*** 0.00*** 0.01***
P-value ANOVA statistic 0.00*** 0.00*** 0.00*** 0.00*** 0.55
Panel B: Parental financial teaching received in different age categories
Panel E: Presence of habit formation as mechanism (expenditures were monitored)
Parental financial teaching only
received in young childhood 0.57 3,178.53 2.97 0.24 3,914.31
I could decide on all of my
expenditures 0.61 3,358 3.10 0.28 4,772
Parental financial teaching only
received in adolescence 0.54 2,105.15 2.95 0.30 2,698.14
I could mostly decide how I
spent my money 0.73 4,388 3.31 0.26 3,395
Parental financial teaching received
throughout childhood 0.72 4,340.42 3.27 0.22 3,355.04
Part of my expenditure was
decided by me, the rest by
my parents 0.71 4,139 3.27 0.25 3,968
P-value ANOVA statistic 0.00*** 0.00*** 0.00*** 0.00*** 0.9254
My parents decided how I
spent most of my money 0.72 4,423 3.27 0.22 4,072
My parents decided how I
spent all of my money 0.68 4,101 3.16 0.16 2,255
Panel C: Presence of modeling as mechanism (received stimulation to save)
P-value ANOVA statistic 0.00*** 0.00*** 0.00*** 0.00*** 0.09*
No, not at all 0.59 3,436 3.00 0.25 3,668
Yes, but only to a certain extent 0.68 3,993 3.18 0.25 4,836
Panel F: Presence of independence as mechanism (received pocket money)
Yes, they told me how important
saving is 0.71 4,165 3.27 0.21 2,839
No 0.66 3,689 3.13 0.16 2,427
Yes, they emphasized the necessity of
saving 0.75 4,677 3.36 0.18 2,906
Occasionally 0.70 3,847 3.16 0.22 3,413
P-value ANOVA statistic 0.00*** 0.00*** 0.00*** 0.00*** 0.09*
Yes, but it was sometimes
forgotten 0.65 3,629 3.16 0.25 3,787
Yes 0.72 4,594 3.33 0.26 4,086
P-value ANOVA statistic 0.00*** 0.00*** 0.00*** 0.00*** 0.08*
The influence of parental financial teaching on saving and borrowing behavior
20
Panel D shows that receiving financial advice from parents during childhood leads at a
minimum to an increase of 8% in willingness to save and almost €250 more in savings. It also
leads to both an increased willingness to put money aside in the future and fewer loans. In
contrast, the amount of debt is highest in the group that received the most financial advice
from parents during childhood. However, this difference is not significant according the
ANOVA test.
Next, panel E indicates that moderate parental supervision on children’s spending
patterns leads to the highest willingness to save and greatest saving amounts in adulthood.
However, individuals whose expenditures were most highly monitored by their parents during
childhood have the lowest number of both loans and debts. These findings therefore show that
parental expenditure monitoring during childhood has an effect on both saving and borrowing
behavior, regardless of the degree of monitoring.
The results presented in panel F show that receiving pocket money in childhood leads
to more willingness to save and higher savings in adulthood. Individuals who consistently
received pocket money have approximately €1,000 more savings than adults who never
received a childhood allowance. Receiving pocket money during childhood leads to more
loans and higher debts in adulthood. Adults who never received pocket money have 10%
fewer loans than those who received it on a regular basis.
4.2. Multivariate results
The multivariate results of study encompass saving and borrowing behavior, as
presented below.
4.2.1. Saving behavior
Table III reports the influence of parents on the saving behavior of adults. Panel A of
this table presents the influence of parental financial teaching on saving behavior. The
willingness to save and total savings relative to net income increase by 2% when individuals
receive parental financial teaching to the highest degree as opposed to not receiving any (at
the 1% statistical level). Total savings and the willingness to save in the future are also
significantly larger (at the 1% statistical level) for individuals who received parental financial
teaching. These findings support hypothesis 1a, namely parental financial teaching influences
the saving behavior of adults. Specifically, parental financial teaching leads to a higher
The influence of parental financial teaching on saving and borrowing behavior
21
willingness to save and more savings. This is in line with the study of Bucciol and Veronesi
(2014), who also find that parental financial teaching leads to more savings. The findings of
the current study that parents influence the formation of their children’s financial behavior,
support both the planned behavior theory and the family systems theory.
Panel B provides evidence that prolonged parental financial teaching has the largest
influence on the saving behavior of adults. All facets of saving behavior are significantly
positive (at a 1% or 5% statistical level) when parental financial teaching is received
throughout childhood. For instance, individuals who receive parental financial teaching in
both age categories are 18% more willing to save than adults who never received such
teaching. The results further indicate (at the 10% statistical level) that when parents only teach
their children during young childhood, only total savings are significantly influenced.
However, this effect is larger when parental financial teaching is also provided during
adolescence. It is also noteworthy that only receiving parental financial teaching during
adolescence has no impact on saving behavior. Teaching children during their adolescence
therefore makes no sense when parental financial teaching has not been provided in previous
years. The above results indicate that the largest effect on saving behavior is achieved when
parents provide their children with financial education during both childhood and
adolescence. Consequently, hypothesis 2a is accepted. This is in line with the study of
Bucciol and Veronesi (2014), who also find evidence that the most effective strategy is to
give children parental financial teaching throughout childhood.
Next, panel C reports the results of being encouraged to start saving during childhood
on saving behavior in adulthood. Individuals are 4% more willing to save when they are
frequently stimulated during childhood to start saving than when they receive no form of
stimulation (at the 1% statistical level). Total savings, future savings and savings relative to
net income are also larger for adults who are stimulated in childhood to save (at the 1%
statistical level). Since parental stimulation is a proxy of parents’ modeling role in this study,
these findings support hypothesis 3a on the role of modeling as mechanism. Parents thus
influence the saving behavior of their children by being a role model. In particular, parents
providing examples to their children, by stimulating particular saving behavior, results in a
higher willingness to save and more savings.
The influence of parental financial teaching on saving and borrowing behavior
22
Table III: Multivariate results, saving behavior
This table presents the multivariate results of the effects that parental financial teaching in childhood
(panel A), parental financial teaching in different age categories (panel B) and different mechanisms
(panel C-F) have on the following facets of saving behavior: willingness in the past to save (column 1), amount of savings per year (column 2), willingness to save in future (column 3), and total savings relative
to net income (column 4). *,**,*** denote significance at the 10%, 5% and 1% level respectively, P-value
based on robust standard errors in parentheses.
Willingness to save
Total
savings per
year (logarithm)
Willingness to save in future
Savings to net income
(1) (2) (3) (4)
Panel A: Parental financial teaching received
Parental financial teaching received 0.02*** 0.17*** 0.04*** 0.02***
(0.00) (0.00) (0.00) (0.00)
Female -0.03* -1.08*** 0.01 -0.11***
(0.10) (0.00) (0.82) (0.00)
Age (in years) -0.00** -0.01 -0.01*** -0.00
(0.01) (0.43) (0.00) (0.26)
Retired (age >65) 0.06*** 0.31 0.14*** 0.04*
(0.00) (0.10) (0.00) (0.06)
High education 0.08* 1.28*** 0.16* 0.12***
(0.09) (0.00) (0.07) (0.00)
Middle education 0.03 0.23 0.04 0.03
(0.59) (0.61) (0.67) (0.54)
Employee 0.10*** 1.06*** 0.20*** 0.10***
(0.00) (0.00) (0.00) (0.00)
Self-employee 0.00 0.46 0.12* 0.05
(0.92) (0.27) (0.08) (0.25)
Participation in a partnership 0.18** 2.02** 0.37** 0.19**
(0.03) (0.01) (0.01) (0.01)
Risk attitude 1 0.01** 0.04 0.02** 0.00
(0.04) (0.17) (0.03) (0.16)
Risk attitude 2 0.01* 0.01 0.02** 0.00
(0.08) (0.83) (0.01) (0.73)
Number of children -0.04** -0.87*** -0.05 -0.09***
(0.02) (0.00) (0.15) (0.00)
Size of the household 0.01 0.43*** 0.03 0.04**
(0.60) (0.01) (0.41) (0.01)
Net income (logarithm) 0.02*** 0.52*** 0.03** 0.03**
(0.00) (0.00) (0.02) (0.01)
Area fixed effects YES YES YES YES
Year fixed effects YES YES YES YES
Constant 0.12 -4.28*** 2.45*** -0.18
(0.31) (0.00) (0.00) (0.17)
R² 0.054 0.109 0.085 0.088
N 8964 8836 8863 8825
The influence of parental financial teaching on saving and borrowing behavior
23
Continued table III: Multivariate results, saving behavior
This table presents the multivariate results of the effects that parental financial teaching in childhood (panel A), parental financial teaching in different age categories (panel B) and different mechanisms (panel C-F) have on
the following facets of saving behavior: willingness in the past to save (column 1), amount of savings per year (column 2), willingness to save in future (column 3), and total savings relative to net income (column 4). *,**,*** denote significance at the 10%, 5% and 1% level respectively, P-value based on robust standard errors in parentheses.
Willingness to save
Total savings
per year (logarithm)
Willingness to save in future
Savings to net income
Willingness to save
Total savings
per year (logarithm)
Willingness to save in future
Savings to net income
(1) (2) (3) (4) (1) (2) (3) (4)
Panel B: Parental teaching received in different age categories
Panel D: Presence of discussion/guidance as mechanism (received financial advice)
Parental financial teaching only received in young childhood 0.08 0.83* 0.10 0.08
Received financial advice 0.03*** 0.25*** 0.06*** 0.02***
(0.16) (0.09) (0.33) (0.10)
(0.00) (0.00) (0.00) (0.00)
Parental financial teaching only received in adolescence 0.01 0.31 0.00 0.03
R² 0.050 0.104 0.082 0.083
(0.90) (0.68) (1.00) (0.67)
Area fixed effects YES YES YES YES
Parental financial teaching
received throughout childhood 0.18*** 1.62*** 0.24** 0.16***
Year fixed effects YES YES YES YES
(0.00) (0.00) (0.01) (0.00)
Socio-demographic
controls YES YES YES YES
R² 0.055 0.107 0.085 0.087
Number of observations 9362 9225 9256 9213
Area fixed effects YES YES YES YES
Year fixed effects YES YES YES YES
Panel E: Presence of habit formation as mechanism (expenditures were monitored)
Socio-demographic controls YES YES YES YES
Expenditures were monitored 0.01** 0.13** 0.03** 0.01**
Number of observations 9569 9425 9457 9412
(0.03) (0.02) (0.02) (0.02)
R² 0.047 0.101 0.081 0.081
Panel C: Presence of modeling as mechanism (received stimulation to save)
Area fixed effects YES YES YES YES
Stimulated to save 0.04*** 0.36*** 0.08*** 0.04***
Year fixed effects YES YES YES YES
(0.00) (0.00) (0.00) (0.00)
Socio-demographic
controls YES YES YES YES
R² 0.052 0.105 0.084 0.086
Number of observations 9406 9270 9295 9259
Area fixed effects YES YES YES YES Year fixed effects YES YES YES YES
Panel F: Presence of independence as mechanism (received pocket money)
Socio-demographic controls YES YES YES YES
Received pocket money 0.00 0.06 0.01 0.01
Number of observations 9461 9320 9349 9308
(0.46) (0.35) (0.48) (0.33)
R² 0.045 0.098 0.080 0.079
Area fixed effects YES YES YES YES
Year fixed effects YES YES YES YES
Socio-demographic controls YES YES YES YES
Number of observations 9576 9431 9463 9418
The influence of parental financial teaching on saving and borrowing behavior
24
Panel D shows that giving children financial advice in budgeting significantly
influences all measured facets of saving behavior (at the 1% statistical level). Receiving
financial advice during childhood leads to a higher willingness to save and more savings in
adulthood. Moreover, adults who received the most parental advice on budgeting during
childhood have 2% more savings relative to their net income than individuals who never
received such advice in childhood. The findings of this study thus support hypothesis 3a,
namely that the guidance and discussion mechanism is one way in which parents influence the
saving behavior of their children. This is in line with the findings of Bucciol and Veronesi
(2014) and Webley and Nyhus (2012).
Panel E then presents the results of controlled spending patterns in childhood on
individual saving behavior. Parental expenditure monitoring has an increasing effect on all
observed types of saving behavior (at the 5% statistical level). Specifically, the willingness to
save among adults increases by 1% when parents control all expenses in childhood as
opposed to undertaking no monitoring. Since monitoring the expenditures of their children is
a method for parents to influence habit formation, the findings of this research support
hypothesis 3a on the habit formation mechanism. By influencing the habit formation of their
children, parents have an impact on the saving behavior of adults.
The influence of receiving pocket money during childhood on saving behavior is
shown in panel F. The saving behavior of adults who received pocket money is not
significantly different from those who never did. Since pocket money is a proxy for being
independent in childhood, hypothesis 3a is rejected for the independence mechanism. Adults
who had more financial responsibility in childhood thus do not behave financially differently
than others. Although the hypothesis cannot be accepted, the findings of this study are partly
supported by the research of Webley and Nyhus (2006), which states that independence is not
the most relevant mechanism for explaining the effect of parental education.
Furthermore, the control variables on average have the expected sign as in previous
studies; please refer to panel A for the exact results of the control variables. As expected,
women have lower savings than men, higher education levels lead to higher savings, adults
with a higher net income save more, risk-tolerant individuals are less willing to save,
employees save more, and households with more children have lower savings. However, the
influence of the number of household members on saving behavior is not as expected,
because this study shows that having more household members leads to more savings. A
possible explanation for this unexpected observation is that larger households likely have
more income to spend and therefore save more. If this explanation is correct, the size of the
The influence of parental financial teaching on saving and borrowing behavior
25
household should have no effect on the savings amount relative to net income, since this
dependent variable controls for changes in net income. However, in this study households
with more than one member save significantly more relative to their net income than single
households. Another explanation is that larger households have more children and therefore
save more with regard to future expenses (e.g. for education). However, this explanation is
also not supported by the observations of this study, as the number of children leads to
significantly lower savings. The finding that larger households save more can therefore not be
explained by this study. It is also noteworthy that individuals who are retired (here adults
above the age of 65) have a higher willingness to save than people who are younger than 65.
The reason for this unexpected finding may be methodological in nature: it could be that most
people in the sample entered retirement after age 65. It is likely that individuals nearing
retirement are more willing to save to cope with their upcoming drop in income.
4.2.2. Borrowing behavior
Table IV presents the influence of parents on the borrowing behavior of individuals.
Panel A shows that adults who received the most parental financial teaching have 1% fewer
loans than those who never received such teaching (at the 1% statistical level). The amount of
debt is also accordingly lower as parental financial teaching is provided (at the 1% statistical
level). Hypothesis 1b is consequently accepted. This study supports the view that parental
financial teaching influences borrowing behavior. This conclusion is in line with the study of
Norvilitis and MacLean (2010) and the planned behavior and family systems theories.
In contrast, panel B provides no evidence for hypothesis 2b. Receiving prolonged
parental financial teaching has no significant effect on the borrowing behavior of adults.
Furthermore, receiving such teaching in only one age category has no effect on the number of
loans and debts. Although parental financial teaching has an effect on borrowing behavior, it
does not matter in which age category it is received.
The influence of parental financial teaching on saving and borrowing behavior
26
Table IV: Multivariate results, borrowing behavior
This table presents the multivariate results of the effects that parental financial teaching in childhood (panel A),
parental financial teaching in different age categories (panel B) and different mechanisms (panel C-F) have on the following facets of borrowing behavior: presence of a loan (column 1), total debt (column 2), and total debt
relative to net income (column 3). *,**,*** denote significance at the 10%, 5% and 1% level respectively, P-
value based on robust standard errors in parentheses.
Presence of a loan
Total debt (logarithm) Total debt to net income
(1) (2) (3)
Panel A: Parental financial teaching received
Parental financial teaching received -0.01*** -0.09*** -0.01***
(0.00) (0.00) (0.00)
Female -0.03* -0.30* -0.03*
(0.10) (0.09) (0.08)
Age (in years) -0.01*** -0.05*** -0.01***
(0.00) (0.00) (0.00)
Retired (age >65) 0.02 0.04 0.00
(0.27) (0.77) (0.87)
High education 0.01 0.18 0.02
(0.90) (0.56) (0.58)
Middle education -0.02 0.03 0.00
(0.74) (0.91) (0.95)
Employee -0.02 -0.12 -0.01
(0.30) (0.37) (0.30)
Self-employee 0.04 0.23 0.02
(0.31) (0.49) (0.56)
Participation in a partnership 0.05 1.20 0.11
(0.44) (0.19) (0.19)
Risk attitude 1 -0.00 -0.05* -0.01*
(0.34) (0.07) (0.08)
Risk attitude 2 -0.00 -0.01 -0.00
(0.20) (0.87) (0.85)
Number of children 0.01 0.00 0.00
(0.55) (0.98) (0.88)
Size of the household -0.02 -0.17 -0.02
(0.16) (0.29) (0.22)
Net income (logarithm) -0.01 -0.04 -0.01*
(0.12) (0.40) (0.07)
Area fixed effects YES YES YES
Year fixed effects YES YES YES
Constant 1.01*** 6.90*** 0.78***
(0.00) (0.00) (0.00)
R² 0.065 0.054 0.054
Number of observations 8970 7000 6991
The influence of parental financial teaching on saving and borrowing behavior
27
Continued table IV: Multivariate results, borrowing behavior
This table presents the multivariate results of the effects that parental financial teaching in childhood (panel A), parental financial teaching in different age categories (panel B) and different mechanisms (panel C-F) have
on the following facets of borrowing behavior: presence of a loan (column 1), total debt (column 2), and total debt relative to net income (column 3). *,**,*** denote significance at the 10%, 5% and 1% level respectively,
P-value based on robust standard errors in parentheses.
Presence of a loan Total debt (logarithm) Total debt to net income (%) Presence of a loan
Total debt (logarithm)
Total debt to net income (%)
(1) (2) (3) (1) (2) (3)
Panel B: Parental teaching received in different age categories
Panel D: Presence of discussion/guidance as mechanism (received financial advice)
Parental financial teaching only
received in young childhood -0.01 0.05 0.00
Received financial advice -0.02*** -0.13** -0.01**
(0.80) (0.91) (0.97)
(0.01) (0.03) (0.03)
Parental financial teaching only received in adolescence 0.08 0.58 0.05
R² 0.061 0.052 0.052
(0.39) (0.45) (0.48)
Area fixed effects YES YES YES
Parental financial teaching
received throughout childhood -0.05 -0.26 -0.03
Year fixed effects YES YES YES
(0.27) (0.52) (0.49)
Socio-demographic controls YES YES YES
R² 0.064 0.054 0.054
Number of observations 9368 7269 7259
Area fixed effects YES YES YES
Year fixed effects YES YES YES
Panel E: Presence of habit formation as mechanism (expenditures were monitored)
Socio-demographic controls YES YES YES
Expenditures were monitored -0.01** -0.08 -0.01
Number of observations 9575 7403 7392
(0.03) (0.11) (0.10)
R² 0.066 0.056 0.056
Panel C: Presence of modeling as mechanism (receiving stimulation to save in childhood)
Area fixed effects YES YES YES
Stimulated to save -0.03*** -0.19*** -0.02***
Year fixed effects YES YES YES
(0.00) (0.00) (0.00)
Socio-demographic controls YES YES YES
R² 0.066 0.055 0.055
Number of observations 9412 7300 7291
Area fixed effects YES YES YES
Year fixed effects YES YES YES
Panel F: Presence of independence as mechanism (received pocket money)
Socio-demographic controls YES YES YES
Received pocket money 0.00 0.01 0.00
Number of observations 9467 7328 7318
(0.89) (0.86) (0.88)
R² 0.061 0.053 0.053
Area fixed effects YES YES YES
Year fixed effects YES YES YES
Socio-demographic controls YES YES YES
Number of observations 9582 7412 7401
The influence of parental financial teaching on saving and borrowing behavior
28
Panel C shows that children whose parents encouraged them to save have fewer loans
and lower debts in adulthood (at the 1% statistical level) compared with adults who were not
stimulated to save during childhood. In addition, the fraction of debt relative to net income is
smaller (at the 1% statistical level) for individuals who were stimulated to start saving during
childhood. Hypothesis 3b is therefore accepted for the modeling mechanism. Modeling is thus
a mechanism that explains parental influence on the borrowing behavior of adults. Although
the measure of modeling in this case relates only to savings, it is likely that the finding is
plausible. Adults who were stimulated to save during childhood have higher savings (please
refer to Section 4.2.1) and therefore less need to borrow money. Furthermore, it is likely that
when parents stimulate their children to save, they also stimulate other financial behavior.
Consequently, parents also impact borrowing behavior by serving as role models and
stimulating their children.
Next, panel D reports the effect that giving financial advice on budgeting to children
has on borrowing behavior in adulthood. Giving such advice to children leads to lower debt in
adulthood (at the 5% statistical level). In addition, individuals who received advice during
childhood to a large extent have 2% fewer loans in adulthood than those who never received
parental advice (at the 1% statistical level). These findings therefore prove that parents
influence the borrowing behavior of individuals through guidance and discussion. Hypothesis
3b is thus accepted for the guidance and discussion mechanism. This is in line with the
research of Norvilitis and MacLean (2010), who find that teaching children how to manage
money leads to lower debt.
Panel E shows the effect of parental control of childhood spending on borrowing
behavior. The results indicate that only loan existence is influenced by parental monitoring
roles (at the 5% statistical level). The amount of debt is not affected by the freedom adults had
to spend their money as they pleased as children. The results therefore provide partial
evidence that habit formation is a mechanism that explains the influence of parents on the
future borrowing behavior of their children. Hypothesis 3b is thus partly accepted for the
habit formation mechanism.
As with saving behavior, independence is also not a mechanism that explains the role
of parents on the financial behavior of their children in relation to borrowing. As shown in
panel F, receiving pocket money during childhood has no influence on the number of loans or
the debts of adults. Hypothesis 3b, about the independence mechanism, is therefore rejected.
The control variables are in line with the study of Webley and Nyhus (2012); please
refer to panel A for the exact results of the control variables. Thus, older persons have fewer
The influence of parental financial teaching on saving and borrowing behavior
29
loans and lower debts. An explanation for this finding is that older persons face more
difficulties in obtaining loans than young adults. Older adults also have had more time to
repay loans that they accumulated during their younger years. The significant results
concerning risk attitude and borrowing behavior have not been observed in previous studies.
An explanation of why risk attitude matters for borrowing behavior is that risk-tolerant
individuals prefer to invest in risky investments (please refer to Section 3.2.3). Loans can be
regarded as risky. However, not all loans are subject to high risk, since some carry low
interest and have long repayment periods. Surprisingly, is that the other variable that
measures risk attitude is not significant, although both variables measure individual attitudes
to risk. It is therefore hard to conclude that risk attitude influences individual borrowing
behavior. The results also show that women have lower numbers of loans and debts than men.
This has not been discovered in previous studies. A possible explanation is that females take
fewer risks than their male counterparts (Eckel and Grossman, 2008). Noteworthy is that my
findings about the control variables are not in line with the study of Nyhus and Webley
(2001). An explanation is that they use surveys of the DHS before the year 2000. According
to Bucciol and Veronesi (2014), surveys before 2000 were primarily completed by wealthier
households. The present study is therefore not comparable with the study of Nyhus and
Webley (2001).
4.3. Additional test for the effect of parental financial teaching on borrowing behavior
Although the different tests for borrowing behavior show that parental financial
teaching leads to fewer loans and less debt, it is hard to state that this is in general positive
behavior. Borrowing money is not always a sign of bad financial behavior. As mentioned in
the introduction, individuals borrow money for instance to make stable and expensive
purchases and to protect themselves against unexpected events. Borrowing to pay tuition fees
is also not a sign of bad borrowing behavior, since it is likely that future income is reserved in
order to repay the loan within the specified payment terms. However, using credit from mail-
order companies or credit cards is generally a poor form of borrowing behavior, seeing as the
interest on these types of loans is often high and the terms and conditions are unfavorable.
In order to determine whether parental financial teaching reduces bad borrowing
behavior, I conduct an additional test. Table V presents the univariate results of receiving
parental financial teaching on the existence and debt amount of related to credit from mail-
order companies and credit cards, since both are common examples of bad loan forms.
The influence of parental financial teaching on saving and borrowing behavior
30
Table V: Additional test, borrowing behavior
This table presents the univariate results of the effect that parental financial teaching in childhood has on following facets of
borrowing behavior: Existence of credits from mail-order companies (column 1), debt amount on credits from mail-order
companies (column 2), existence of credit cards (3), and debt amount on credit cards(column 4). *,**,*** denote significance on the ANOVA test at the 10%, 5% and 1% level respectively.
Parental financial teaching received
(divided in quartiles)
Existence of credits from
mail-order companies
Debt amount on credits from mail-
order companies
Existence of credit
cards
Debt amount on
credit cards
(1) (2) (3) (4)
1 (low) 0.02 22 0.04 74
2 0.01 6 0.04 41
3 0.02 16 0.05 72
4 (high) 0.00 1 0.03 10
P-value ANOVA test 0.00*** 0.00*** 0.00*** 0.00***
For the individuals in the highest quartile of parental financial teaching, the existence
of both loans and the corresponding debt shields are significantly lower compared to the other
quartiles (according the ANOVA test). Moreover, the group that receives a minimum of
parental financial teaching has the highest number of credits from mail-order companies.
These individuals also have the highest amount of debt on both type of loans. Consequently,
this additional test shows that parental financial teaching reduces individuals’ bad borrowing
behavior.
5. Conclusion and discussion
Several previous studies have evaluated the role of financial education in financial
behavior, in order to improve the financial decisions of households. This paper investigates
the role that parents play in the formation of their children’s saving and borrowing behavior.
Based on the results I conclude that receiving parental financial teaching leads to a higher
willingness to save, more savings, fewer loans and lower debts in adulthood. Both the planned
behavior and family systems theories support the finding that parents influence the saving and
borrowing behavior of their children.
Specifically, I find that the effect of parental financial teaching on saving behavior is
the largest when children receive this teaching during both their young childhood and their
adolescence. This is logical, since children who received prolonged financial education come
more frequently into contact with the advantages of parental financial teaching. However, this
result is not without biases. The measurement method used do not exclude the presence of
other forms of parental teaching in the observed age groups, since the data do not provide
The influence of parental financial teaching on saving and borrowing behavior
31
sufficient information about the precise age at which parental financial teaching is received.
Future research should therefore include data about the exact age at which children receive
parental financial teaching in order to present more representative conclusions.
Additionally, this research offers proof that modeling, guidance and discussion, and
habit formation are mechanisms that explain the manner in which parents influence the
formation of their children’s saving and borrowing behavior. In the case of borrowing
behavior, my findings show that parents have more impact as models and advisory agents
than as influencers in the formation of habits. This is in line with Webley and Nyhus (2006),
who state that modeling and discussion/guidance are the most relevant mechanisms. A
possible explanation is that these mechanisms addressed financial behavior directly and not in
an indirect way such as by looking at the habit formation and independence mechanisms.
However, this research does not describe the full effect of the mechanisms, since parents have
more possibilities to be involved in the different mechanisms than covered in this study. For
instance, pocket money is not the only tool that parents have to give children financial
responsibilities. Future research should therefore include more information about parent’s
attempts to make children more financially aware, in order to better observe the effects of the
four mechanisms.
This study has a number of limitations, each of which points the way to future
research. First, it is likely that there are more mechanisms behind the effects that parents have
on their children’s saving and borrowing behavior than investigated here. For instance,
Cronqvist and Siegel (2015) find that genetic differences between individuals explain one
third of the variation in saving behavior. Future research should thus include heritability in
order to obtain a better and more representative view of why parents influence the saving and
borrowing behavior of their children. Second, it is hard to state on basis of my findings
whether parental financial teaching has positive or negative effects on borrowing behavior,
since not all types of loans represent bad borrowing behavior. Although my additional test on
borrowing behavior indicates that parental financial teaching reduces bad borrowing behavior,
future research should investigate this in more detail to achieve clear and distinct conclusions.
Third, the results are not controlled for other forms of financial education. For instance,
previous studies have observed that providing individuals with financial education in the
workplace influences saving or borrowing behavior (Bayer et al., 2009). It would therefore be
interesting for future research to study the benefit of parental teaching in comparison to other
types of financial education. Fourth, while only the heads of households are observed here, it
The influence of parental financial teaching on saving and borrowing behavior
32
is likely that other household members also influence financial decisions. Without
information on other household members, the results of my study are thus biased.
Despite these limitations, this study adds to the existing literature concerning parental
financial teaching on the saving and borrowing behavior of adults. First, it confirms existing
findings in the area of parental financial teaching. Second, it provides new insights into the
influence of receiving parental teaching throughout childhood and the effect of parental
financial teaching on reducing bad borrowing behavior. Third, it outlines how parents
influence the future saving and borrowing behavior of their children.
Furthermore, the results are also useful and relevant for policy makers. The indication
that parents plays an important role in the formation of children’s financial behavior means
than policy makers should stimulate parents to provide their children with financial teaching.
They could achieve this goal by arming parents with information on how to familiarize their
children with making financial decisions.
In conclusion, parents influence the formation of their children’s saving and borrowing
behavior. However, further research on the topic is needed for more representative results and
explanations.
The influence of parental financial teaching on saving and borrowing behavior
33
6. References
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human
Decision Processes, 50(2), 179-211.
Alessie, R., van Rooij, M. and Lusardi, A. (2011). Financial literacy and retirement
preparation in the Netherlands. Journal of Pension Economics and Finance, 10(4), 527-545.
Ashby, J.S., Schoon, I. and Webley, P. (2011). Save now save later? Linkages between saving
behavior in adolescence and adulthood. European Psychologist, 16, 227-237.
Batty, M., Collins, J. M. and Odders-White E. (2015). Experimental evidence on the effects of
financial education on elementary school students’ knowledge, behavior and attitudes.
Journal of Consumer Affairs, 49(1), 69-96.
Bayer, P.J., Bernheim, D.B. and Scholz, J.K. (2009). The effects of financial education in the
workplace: evidence from a survey of employers. Economic Inquiry, 47(4), 605-624.
Benton, M., Meier, S. and Sprenger, C. (2007). Overborrowing and undersaving: lessons and
policy implications from research in behavioral economics. Public and Community Affairs
discussion paper. Available at:
https://www.bostonfed.org/commdev/pcadp/2007/pcadp0704.pdf.
Bernheim, D.B., Garett, D.M. and Maki, D.M. (2000). Education and saving: the long-term
effects of high school financial curriculum mandates. Journal of Public Economics, 80, 435-
465.
Boonen, A. (2015). Vermogen maakt verschil: de verdeling van vermogen en de gevolgen
ervan. Available at:
http://www.deburcht.nl/userfiles/file/Vermogen%20maakt%20verschil(2).pdf.
Breusch, T.S. and Pagan, A.R. (1980). The lagrange multiplier test and its applications to
model specification in econometrics. The Review of Economic Studies, 47(1), 239-253.
Bucciol, A. and Veronesi, M. (2014). Teaching children to save: what is the best strategy for
lifetime savings? Journal of Economic Psychology, 45, 1-17.
The influence of parental financial teaching on saving and borrowing behavior
34
Bucks, B. and Pence, K. (2006). Do homeowners know their house values and mortgage
terms? Working paper, Federal Reserve Board of Governors. Available at:
http://www.federalreserve.gov/pubs/feds/2006/200603/200603abs.html.
Cronqvist, H. and Siegel, S. (2015). The origins of savings behavior. Journal of Political
Economy, 123(1), 123-169.
Eckel, C. and Grossman, P. (2008). Men, women and risk aversion: experimental evidence.
Handbook of Experimental Economics Results, 1, 1061-1073.
Hira, T.K. and Mugenda, O. (2000). Gender differences in financial perceptions, behaviors
and satisfaction. The Journal of Financial Planning, 13(2), 86-92.
Jacobs-Lawson, J. and Hershey, D. (2005). Influence of future time perspective, financial
knowledge, and financial risk tolerance on retirement saving behaviors. Financial Services
Review, 14(4), 331-344.
Lusardi, A. and Mitchell, O. (2007b). Financial literacy and retirement preparedness:
evidence and implications for financial education. Business Economics, 42(1), 35–44.
Lusardi, A. and Mitchell, O. (2014). The economic importance of financial literacy: Theory
and evidence. Journal of Economic Literature, 52(1),5-44.
Moore, T.J. and Asay, S.M. (2013). Family resource management. Second edition. Los
Angeles: Sage publications.
Norvilitis, J.M. and MacLean, G.M. (2010). The role of parents in college students’ financial
behaviors and attitudes. Journal of Economic Psychology, 31(1), 55-63.
Nyhus, E. K. and Webley, P. (2001). The role of personality in household saving and
borrowing behavior. European Journal of Personality, 15, 85-103.
Petersen, M.A. (2009). Estimating standard errors in finance panel data sets: comparing
approaches. Review of Financial Studies, 22(1), 435-480.
Shim, S., Barber, B. L., Card, N. A., Xiao, J.J. and Serido, J. (2010). Financial socialization of
first-year college students: the roles of parents, work and education. Journal of Youth and
Adolescence, 39, 1457-1470.
The influence of parental financial teaching on saving and borrowing behavior
35
Teppa, F. and Vis, C. (2012). The centERpanel and the DNB household survey:
methodological aspects. DNB Occasional Studies, 10(4).
Torres-Reyna, O. (2007). Panel data analysis fixed and random effects using stata. [Online]
available at: http://www.princeton.edu/~otorres/Panel101.pdf.
Vassallo, A.C.R. (2003). Children’s perception of their influence over purchases: the role of
parental communication patterns. Journal of Consumer Marketing, 20(1), 55-66.
Von Bertalanffy, L. (1968). General system theory: Foundations, Development, Applications.
New York: George Braziller.
Webley, P. and Nyhus, E.K. (2006). Parents’ influence on children’s future orientation and
saving. Journal of Economic Psychology, 27(1), 140-164.
Webley, P. and Nyhus, E.K. (2012). Economic socialization, saving and assets in European
young adults. Economics of Education Review, 33, 19-30.
Willis, L.E. (2011). The Financial Education Fallacy. American Economic Review, 101(3),
429–434.
The influence of parental financial teaching on saving and borrowing behavior
36
Appendix A: Survey questions and construction of key variables
Saving behavior - Willingness to save: Dummy variable indicating [1] when respondents answered: “Yes” to the following question: “Did your household put any money aside in the past 12 months?” and [0] otherwise.
- Total savings: The amount of saving per year. Derived from the following question: “How much money did your household put
aside in the past 12 months?”, with a discrete answer scale. This scale have seven tiers between 0 and more than 75,000 Euros. The continuous variable is set equal to the central value of each range. For the extreme ranges, the threshold value is used to
create a continuous variable.
- Willingness to save in future: Ordinal variable indicating: [1] when respondents answered: “No, certainly not”, [2] when respondents answered: “No, probably not”, [3] when respondents answered: “Yes, perhaps”, and [4] when respondents answered:
“Yes, certainly” to the following question: “Is your household planning to put money aside in the next 12 months?”
- Total savings to net income: Measured as total savings per year divided by net income. Borrowing behavior - Presence of a loan: Dummy variable indicating [1] when respondents answered: “Yes” to the following question: “Did you, on
31 December <year>, have one or more <type of loan>?” and [0] otherwise. The type of loan is one of the following: private
loans; extended lines of credit; credits by mail-order companies; finance credit; loans from family, friends, or acquaintances;
study loans; credit cards; and other loans. - Total debt: The amount of total debt on the different loans of the respondents. Derived from the following question: “What was
the remaining debt on your <type of loan> on 31 December <year>?”
- Total debt to net income: Measured as total debt divided by net income.
Parental financial
teaching
- Parental financial teaching received: The degree in which parental financial teaching is received. It is measured as the total
sum of the ordinal answer scales of the four questions described below, which indicate the presence of different mechanisms. The
value of the variable ranges from 4 to 17.
- Modeling mechanism / received stimulation to save: Ordinal variable indicating: [1] when respondents answered: “No not at
all”, [2] when respondents answered: “Yes, but only to a certain extent”, [3] when respondents answered: “Yes, they told me how
important saving is”, and [4] when respondents answered: “Yes, they emphasized the necessity of saving” to the following question: “Did your (grand)parents stimulate you to save money between the ages of 12 and 16?”
- Discussion and guidance mechanism / received financial advice: Ordinal variable indicating: [1] when respondents answered:
“No”, [2] when respondents answered: “Yes, but only to a certain extent”, [3] when respondents answered: “Yes, they gave me some advice and practical help”, and [4] when respondents answered: “Yes, they gave me advice and practical help” to the
following question: “Did your (grand)parents try to teach you how to budget when you were between 12 and 16 years of age?”
- Habit formation mechanism / expenditures were monitored: Ordinal variable indicating: [1] when respondents answered: “I
could decide on all of my expenditures”, [2] when respondents answered: “I could mostly decide how I spent my money”, [3]
when respondents answered: “Part of my expenditure was decided by myself, the rest by my parents”, [4] when respondent
answered: “My parents decided how I spent most of my money”, and [5] when respondents answered: “My parents decided how I
spent all of my money” to the following question: “When you were between 8 and 12 years of age, could you spend your money
as your pleased?”
- Independence mechanism/ received pocket money: Ordinal variable indicating: [1] when respondents answered: “No”, [2] when respondents answered: “Occasionally”, [3] when respondents answered: “Yes, but it was sometimes forgotten”, and [4]
when respondents answered: “Yes” to the following question: “When you were between 8 and 12 years of age, did you receive an
allowance from your parents? By allowance we mean a fixed amount received on a regular basis”. - Parental financial teaching in young childhood: Dummy variable indicating [1] when respondents only received in young
childhood parental financial teaching, and [0] otherwise. Parental financial teaching is only received in young childhood if the
mechanisms habit formation or/and independence are present and not the mechanisms modeling and/or discussion and guidance. A mechanism is present if the value of its corresponding value differs from 1 (please refer to the four mechanism variables
described above).
- Parental financial teaching in adolescence: Dummy variable indicating [1] when respondents only received parental financial
teaching in adolescence, and [0] otherwise. Parental financial teaching is only received in adolescence if the mechanisms modeling and/or discussion and guidance are present and not the mechanisms habit formation and/or independence. A mechanism
is present if the value of its corresponding value differs from 1 (please refer to the four mechanism variables described above).
- Parental financial teaching throughout childhood: Dummy variable indicating [1] when respondent received throughout childhood parental financial teaching, and [0] otherwise. Parental financial teaching is received throughout childhood if parental
teaching is received in both young childhood (if the mechanisms habit formation and/or independence are present) and
adolescence (if the mechanisms modeling and/or discussion and guidance are present). A mechanism is present if the value of its corresponding value differs from 1 (please refer to the four mechanism variables described above).
Gender Dummy variable indicating [1] if the respondent is female and [0] for male.
Age Age of the respondent.
Retired Dummy variable indicating [1] if age of the respondent is above 65 and [0] otherwise. Occupation Employee: Dummy variable indicating [1] if the respondent works as an employee and [0] otherwise. Self-employee: Dummy variable indicating [1] if the respondent is self-employed and [0] otherwise.
Participation in a partnership: Dummy variable indicating [1] if the respondent participates in a partnership and [0] otherwise.
No work: Dummy variable indicating [1] if the respondent is unemployed and [0] otherwise (included as omitted variable in the
analysis).
Education High education: Dummy variable indicating [1] if the respondent completed vocational college or university education as highest education and [0] otherwise.
Middle education: Dummy variable indicating [1] if the respondent completed pre-vocational education, pre-university
education, or senior vocational training as highest education and [0] otherwise. Low education: Dummy variable indicating [1] if the respondent completed special education, kindergarten/primary school, other
sort of education/training, or no form of education as highest education and [0] otherwise (included as omitted variable in the
analysis).
The influence of parental financial teaching on saving and borrowing behavior
37
Continued appendix A: Survey questions and construction of key variables
Risk attitude Both risk attitude variables ranges from 1 to 7. Where [1] indicates a risk-tolerant individual and [7] a risk-averse individual.
Derived from a factor analysis (please refer to appendix B for the complete factor analysis) based on the degree of agreement on the following statements (statements with an (R) has been reversed to make all the statements consistent) using a seven-point
scale [1="totally disagree" to 7="totally agree"]:
"I think it is more important to have safe investments and guaranteed returns, than to take a risk to have a chance to get the highest
possible returns."
"I do not invest in shares, because I find this too risky."
"If I think an investment will be profitable, I am prepared to borrow money to make this investment" (R)
"I want to be certain that my investments are safe."
"If I want to improve my financial position, I should take financial risks." (R)
"I am prepared to take the risk to lose money, when there is also a chance to gain money." (R)
Number of children Number of children of the respondent.
Size of households Household size of the respondent.
Net income Net income of the respondent. This variable is computed through the DHS itself. In order to create net income, mortgage interest payments and income tax are subtracted from a large range of different sources of income.5
Appendix B: Risk factor analysis
Table VI: Risk factor analysis
Factor analysis of the risk aversion using principal-component factors with a varimax rotation factors. The statement with an (R) has been reversed to
make all the statements consistent. The largest factor loadings in each column are highlighted in bold
Factors Factor 1 Factor 2
Risk attitude 1 -0.08003 0.48955
Risk attitude 2 0.16419 0.27169
Risk attitude 3 (R) 0.37687 -0.11616
Risk attitude 4 -0.09346 0.49864
Risk attitude 5 (R) 0.42404 -0.11761
Risk attitude 6 (R) 0.40858 0.02955
5 In the documentation of the data sets, the calculation of net income is fully described. That can be found at:
http://www.dhsdata.n.l/site/users/login